StyleGAN2-ADA: adaptive discriminator augmentation that helps use StyleGAN2 with small datasets

Released in: Training Generative Adversarial Networks with Limited Data



Training generative adversarial networks (GAN) using too little data typically leads to discriminator overfitting, causing training to diverge. The authors propose an adaptive discriminator augmentation mechanism that significantly stabilizes training in limited data regimes. The approach does not require changes to loss functions or network architectures, and is applicable both when training from scratch and when fine-tuning an existing GAN on another dataset. The paper demonstrates, on several datasets, that good results are now possible using only a few thousand training images, often matching StyleGAN2 results with an order of magnitude fewer images. This can open up new application domains for GANs. The authors also find that the widely used CIFAR-10 is, in fact, a limited data benchmark, and improve the record FID from 5.59 to 2.42.


Year Released

Key Links & Stats



Training Generative Adversarial Networks with Limited Data

@article{DBLP:journals/corr/abs-2006-06676, author = {Tero Karras and Miika Aittala and Janne Hellsten and Samuli Laine and Jaakko Lehtinen and Timo Aila}, title = {Training Generative Adversarial Networks with Limited Data}, journal = {CoRR}, volume = {abs/2006.06676}, year = {2020}, url = {}, eprinttype = {arXiv}, eprint = {2006.06676}, timestamp = {Wed, 17 Jun 2020 14:28:54 +0200}, biburl = {}, bibsource = {dblp computer science bibliography,} }

ML Tasks

  1. General
  2. Image Generation
  3. Style Transfer

ML Platform

  1. Pytorch


  1. Still Image


  1. General
  2. Facial
  3. Digital Human

CG Platform

  1. Not Applicable

Related organizations